Mercuric reductase is a homodimeric protein with two interfacial active sites per dimer. Extensive evidence, obtained with enzyme from a Pseudomonas transposon Tn501, indicates that the active site environments are asymmetric when pyridine nucleotide substrates are bound at both sites, but symmetric in the absence of ligands. Since the complexed enzyme is catalytically relevant, we have proposed a role for the asymmetry in the catalytic mechanism. To elucidate the mechanistic role, we also need to understand how the active sites sense each other. Thus, we want to define both the protein-ligand interactions and the molecular pathway through the protein structure that are involved in the development of asymmetry. Towards this end, we are pursuing experiments to determine which portions of the ligands are essential for asymmetric binding. With that information, we can examine the structure using the MidasPlus software in the graphics lab to identify specific protein-ligand contacts and potential pathways from those contacts that may participate in communication between the active sites. The structural data available is for the enzyme from a Bacillus sp., which is 41% identical in its primary sequence to the Tn501 enzyme. Our major efforts in the graphics lab have been to utilize programs such as MidasPlus, Bloop and Look to construct an homology model of the Tn501 structure based on the Bacillus structure. We have completed our model-building efforts and now primarily use the graphics to examine our structural model to design appropriate experiments to test our hypothesis about site-site communication in this protein.